FACTORS AND STIMULI 211 



and stimuli concerned, and orientation or sense of direction. The first 

 of these is naturally beyond the reach of experiment; in it the influ- 

 ence of the glacial period has probably been much overemphasized, 

 and migration may well have begun earlier. Certainly, the assumption 

 that the climate during middle and late Tertiary was notably warmer 

 and more equable and hence attended with little or no zonation far 

 into the arctic regions is no longer tenable, as the revaluation of the 

 classic fossil floras of North America has shown in particular (Clem- 

 ents, 1916:362; Chaney, 1925; Berry, 1922, 1930). The other two 

 problems are appropriate subjects for both direct and indirect experi- 

 ment, and the positive advances recently made have been obtained 

 from such procedure. In this must be included bird-banding, which 

 has become increasingly definite and experimental since its first use 

 in 1899 (]\Iortensen, 1906), and bids fair to develop into the primary 

 method (cf. Baldwin, 1919; Lincoln, 1924, 1927; Thomson, 1926:143, 

 1936). 



FACTORS AND STIMULI 



Historical. Before considering the experimental studies that bear 

 upon the causes of migration, a helpful background may be afforded 

 by a brief account of views as to the role of light, length of day, and 

 gonad hormones. The opinion that light, as such, furnishes the signal 

 for migration has long been held and often expressed in poetic form 

 (Eifrig, 1924). It was trenchantly criticized by Newton (1874), who 

 pointed out that the theory refuted itself in view of the fact that the 

 southern movement, in particular, was initiated and in large part 

 accomplished at a time when the birds were journeying to increas- 

 ingly shorter days. Nevertheless, it was restated by Seebohm (1888), 

 who expressly eliminated temperature and food to emphasize the 

 search for light as the sole cause. Schafer (1907) seems to have been 

 the first to recognize that light was properly to be interpreted in terms 

 of length of day, and stated that this was important not in itself but 

 only for the purpose of gathering food. Eifrig (1924) has enthusiasti- 

 cally championed the cause of light and day length, and Thomson 

 (1926) regards it as possibly important. Allard appears to give re- 

 strained support to the hypothesis (1928), and Bissonette (1937) in- 

 directly conveys the same impression. Rowan (1926, 1932) at first 

 ascribed much influence to length of day, but later regarded this as 

 largely if not wholly a matter of time for activity and food gathering 

 (cf. Cole, 1933). Heape (1931) concluded that light does not exert a 

 direct effect on migration, but acts in a mediate fashion through vita- 

 mins upon gonads. 



